Project Summary When people explore a new environment, they use at least 2 strategies to travel to areas of interest. The allocentric strategy uses landmarks, while the egocentric strategy uses body position, speed, and time. Initially, use of the landmark strategy activates the hippocampus and use of the body position strategy uses the striatum. Sleep is thought to be important in integrating and distributing hippocampal information to cortical structures, such that there is less hippocampal activation, while at the same time increasing striatal activation upon subsequent experience in the same environment that co-occurs with improved performance, a process known as systems consolidation. Although systems consolidation is one of the major theories of sleep's function in memory, the evidence supporting this derives largely from animals and is lacking in human subjects. Using functional magnetic resonance imaging (fMRI) of the brain during periods of spatial navigation before and after normally consolidated sleep, we can identify changes in hippocampal and striatal activation that support the systems consolidation idea (Aim 1). Sleep disruption has been identified as one of the factors that negatively influence memory formation. Using a continuous positive airway pressure (CPAP) withdrawal model in subjects who have severe obstructive sleep apnea (OSA) and who are fully adherent to CPAP on a nightly basis, we demonstrated that sleep supports the consolidation of spatial navigational memory, and this benefit is significantly attenuated when rapid eye movement (REM) sleep is disrupted by OSA induced exclusively in REM sleep. Additionally, we also demonstrated that aging-associated disruption of slow wave sleep (SWS) in the absence of OSA is associated with impaired consolidation of spatial navigational memory without changes in morning psychomotor vigilance. While the functional outcome on spatial navigational performance is similarly negative when either REM or SWS is disrupted, the brain circuit mechanisms by which this occurs are likely to be different, as REM and SWS differ in several significant ways, including neurotransmitter tone, degree of neuronal synchrony, and frequency of cortical local field potential oscillations. Proof of this hypothesis would be aided by the ability to induce sleep fragmentation alone in a stage specific manner using auditory stimuli within individual subjects without OSA in a way that recapitulates the frequency of sleep fragmentation from severe OSA, thereby controlling for the potentially confounding effect of intermittent hypoxia from OSA (Aim 2). Our observations that disruption of either REM or SWS impairs spatial navigational memory performance suggest that it may do so by disrupting the systems consolidation process. We hypothesize that SWS disruption impairs hippocampal communication with frontal cortex, and as a result, hippocampal activation will remain high following disrupted SWS. Conversely, we predict REM disruption will impair the recruitment of striatum that ordinarily allows a shift to a more egocentric navigation strategy thereby reducing activation of the striatum following disrupted REM (Aim 3).